Download - RedoxModulationofFlavinandTyrosineDetermines ...coupled electron transfer reactions. Photoinduced electron transfer in biological systems, espe-cially in proteins, is a highly intriguing

Redox Modulation of Flavin and Tyrosine DeterminesPhotoinduced Proton-coupled Electron Transfer andPhotoactivation of BLUF Photoreceptors□S

Received for publication, June 14, 2012, and in revised form, July 24, 2012 Published, JBC Papers in Press, July 25, 2012, DOI 10.1074/jbc.M112.391896

Tilo Mathes‡1,2, Ivo H. M. van Stokkum‡, Manuela Stierl§, and John T. M. Kennis‡3

From the ‡Biophysics Group, Department of Physics and Astronomy, Faculty of Sciences, Vrije Universiteit, De Boelelaan 1081A,1081 HV, Amsterdam, The Netherlands and the §Institut fur Biologie, Experimentelle Biophysik, Humboldt Universitat zu Berlin,Invalidenstrasse 42, D-10115 Berlin, Germany

Background: Proton-coupled electron transfer is the key step in BLUF photoactivation.Results: Redox modulation of flavin and tyrosine determines electron transfer rates and signaling efficiency and reveals a newphotocycle intermediate.Conclusion: Partial charge transfer from tyrosine to flavin takes place prior to full electron transfer.Significance:Mechanistic details of protein-modulated electron transfer processes are crucial to understand biological proton-coupled electron transfer reactions.

Photoinduced electron transfer in biological systems, espe-cially in proteins, is a highly intriguing matter. Its mechanisticdetails cannot be addressed by structural data obtained by crys-tallography alone because this provides only static informationon a given redox system. In combination with transient spec-troscopy and site-directed manipulation of the protein, how-ever, a dynamic molecular picture of the ET process may beobtained. In BLUF (blue light sensors using FAD) photorecep-tors, proton-coupled electron transfer between a tyrosine andthe flavin cofactor is the key reaction to switch from a dark-adapted to a light-adapted state, which corresponds to the bio-logical signaling state. Particularly puzzling is the fact that,although the various naturally occurring BLUF domains showlittle difference in the amino acid composition of the flavinbinding pocket, the reaction rates of the forward reaction differquite largely from a few ps up to several hundred ps. In thisstudy, we modified the redox potential of the flavin/tyrosineredox pair by site-directed mutagenesis close to the flavin C2carbonyl and fluorination of the tyrosine, respectively. We pro-vide information onhowchanges in the redox potential of eitherreaction partner significantly influence photoinduced proton-coupled electron transfer. The altered redox potentials allowedus furthermore to experimentally describe an excited statecharge transfer intermediately prior to electron transfer in theBLUFphotocycle. Additionally, we show that the electron trans-fer rate directly correlates with the quantum yield of signalingstate formation.

BLUF (blue light receptors using FAD) proteins are respon-sible for photoadaptive responses of many prokaryotes and afew eukaryotes (1, 2). These blue light induced reactions varyfrom phototaxis (3, 4) and photosynthetic gene regulation (5, 6)in phototrophic organisms to biofilm formation (7) and evenvirulence (8) in pathogenic bacteria. These receptor proteinsaremodularly designed and contain the�150-amino acid largeflavin-binding BLUF domain to modulate the activity of theircorresponding effector domain. In many cases, the effectordomain is directly fused to the receptor. Lately, these BLUF-coupled effectors, predominantly BLUF domain-regulatedenzymes, have been successfully used as so-called optogenetictools to manipulate, for example, the secondmessenger level ina given cell (type) by application of light (9–12). Still, the pho-toactivationmechanism and communication between receptorand effector are not well understood. After excitation with bluelight, proton-coupled electron transfer (PCET)4 from a nearby,conserved tyrosine side chain (Tyr-8 (Y8) in Fig. 1B) to theflavin takes place (13). In the dark-adapted state, this reactionoccurs in a strictly sequential order of electron transfer fol-lowed by proton transfer, whereas in the light-adapted state, ahighly concerted electron and proton transfer reaction leads tothe formation of the same neutral flavin/tyrosine radical pair(13–16). The neutral radical pair recombines to the oxidizedstate, which results in a rearrangement of the hydrogen bondnetwork mainly between flavin, tyrosine, and a conserved glu-tamine residue. The rearranged hydrogen bond network ischaracterized by a downshift of carbonyl vibrations of the fla-vin, thus indicating a stronger hydrogen bond coordination atthis functional group (17, 18). This ismost likely predominantlyfacilitated by the rotation of the glutamine amide side chain (19,20) and leads to the red-shifted absorbance spectrum of thelight-adapted state (5). Additionally, an unusually strong

□S This article contains supplemental Figs. S1–S4.1 Supported by the Chemical Sciences Council of the Netherlands Organi-

zation for Scientific Research through an ECHO grant (to J. T. M. K.), theLaserLab Europe access program (LCVU-1618), and the DeutscheForschungsgemeinschaft.

2 To whom correspondence should be addressed. Tel.: 31-20-59-87935;E-mail: [email protected].

3 Supported by the Chemical Sciences Council of the Netherlands Foundationfor Scientific Research through a Vici grant.

4 The abbreviations used are: PCET, proton-coupled electron transfer; EADS,evolution-associated difference spectra; GSB, ground state bleach; ESA,excited state absorption; SE, stimulated emission; SADS, species-associ-ated difference spectra; CT, charge transfer; ET, electron transfer.

THE JOURNAL OF BIOLOGICAL CHEMISTRY VOL. 287, NO. 38, pp. 31725–31738, September 14, 2012© 2012 by The American Society for Biochemistry and Molecular Biology, Inc. Published in the U.S.A.

SEPTEMBER 14, 2012 • VOLUME 287 • NUMBER 38 JOURNAL OF BIOLOGICAL CHEMISTRY 31725

by guest on July 30, 2020http://w

ww

.jbc.org/D

ownloaded from

http://www.jbc.org/cgi/content/full/M112.391896/DC1

http://www.jbc.org/

hydrogen bond is formed from the conserved tyrosine to theglutamine side chain (19, 21). The molecular details of bothdark- and light-adapted states as well as the implications forsignal transduction are still under heavy debate because theavailable crystal structures of BLUF domains gave contradict-ing results, especially on the orientation of the glutamine sidechain (22–27). Additionally, theoretical calculations indicatethat a tautomerization of the glutamine side chain is anotherpossible hydrogen bond switch mechanism, which is, however,still lacking hard experimental proof (28–31). The hydrogenbond switched stated is also considered to be representative forthe biologically active state. Still, the structural changes (mostlikely induced by the hydrogen bond switch) that facilitate bio-logical signaling by changes in the interaction of the BLUFdomain with other proteins or directly fused effector domainsare very small and poorly understood. So far only one structureof a BLUF-effector complex has been solved and a model forsignal transduction has been established (22). It is very likely,however, that there are various molecular ways of signaling inBLUF photoreceptors similar to what is known about Light-Oxygen-Voltage (LOV) domain signaling (32).In this article, we address the very first reaction after blue

light excitation in the BLUF domain: photoinduced electrontransfer. In addition to their very intriguing signal transductionaspect, BLUF domains are also very powerful model systems tostudy PCET. PCET involving a tyrosine side chain is a key reac-tion in the oxygen-evolving photosystem II complex and up tonow is not fully understood (33–35).Mostmodel studies in thisfield are carried out using small heavymetal complexes ormod-ified cytochrome c or azurin proteins. In contrast to these arti-ficial model systems, BLUF domains present the smallest so farinvestigated evolutionarily optimized model system to studyPCET. In a previous publication (15), we addressed the effectsof hydrogen bonding in dark- and light-adapted states on thesequence of electron and proton transfer. In the light-adaptedstate, we observed a highly efficient proton transfer reactionconcerted with electron transfer within 1 ps. In the dark-adapted state, proton transfer lags electron transfer by (tens of)picoseconds. Thus, the light-adapted state is preconfigured forproton transfer. Here we address the differences in photoin-duced electron transfer between various BLUF domains.Although all BLUF domains are very similar in structure andsequence of the flavin binding pocket, the ultrafast electrontransfer varies from a few ps to hundreds of ps within the BLUFfamily (13, 14, 36–44). Amajor factor that determines electrontransfer in chemistry and biology is the relative redox potentialsof the reaction partners. Previously, Ishikita (45) reported atheoretical study on the redox potential of the one-electronreduction of FMN in flavodoxin and its susceptibility to theprotein environment. In a later study, he addressed the effectsof the hydrogen bond network in AppA on the redox potentialof the reactive tyrosine (46). Accordingly, we observe here thatmodulation of the redox potential of both flavin and tyrosinesignificantly influences the rate of electron transfer. Due tothese altered kinetic properties, the observation of a previouslyoccluded excited state charge transfer intermediate is facili-tated. Selected mutations and modifications were introduced

into Slr1694 (also referred to as SyPixD) from Synechocystis sp.PCC 6803 and studied by ultrafast spectroscopy.

EXPERIMENTAL PROCEDURES

Expression Strain Generation and Characterization—CmpX13(47) was rendered tyrosine-auxotrophic by in-frame deletion oftyrA according to established homologous recombination pro-tocols (48). A linear double-stranded DNA fragment contain-ing a kanamycin resistance conferring cassette flanked by FLPrecombinase recognition target sites was amplified by PCRfrom pKD4 (49), including a 50-bp homology region flankingthe tyrA gene at the 5�- and 3�-end using the primers DtyrA-5�and DtyrA-3� (Table 1). After transformation of the recipientstrain, clones were selected on kanamycin containing LB-agarand colony-purified. Successful disruption of the tyrA gene wasconfirmed by growth tests in minimal media with and withoutthe addition of L-tyrosine. The genomically integrated kanamy-cin resistance-conferring cassette of a single correctly identifiedclone was removed by expression of FLP recombinase accord-ing to protocols described previously (47, 49). The resultingclones were colony-purified and selected for kanamycin sensi-tivity and tyrosine auxotrophy. A single clone, henceforthnamed CpX�Y, was verified by DNA sequence analysis andused for protein production as indicated below.The tyrosine requirement ofCpX�Ywas determined relative

to the glucose consumption. Cells were grown in M9 minimalmedium under glucose-limiting conditions (0.1% (w/v)) withvarying concentrations of L-tyrosine. The cell density was esti-mated by absorbance measurements at 600 nm every 20 minduring growth in a 96-well plate shaking incubator at 37 °C.After reaching the stationary phase, the L-tyrosine minimalrequirement was extracted at concentrations slightly below themaximum growth level in apparent L-tyrosine non-limitingconditions (not shown).Mutants and Protein Production—Slr1694 mutants were

produced from pET28(�)-slr1694, as described previously(44). Site-directed mutations were introduced according to theQuikChangeTM (Stratagene) protocol using the primer pairsN31R/N31R_r and N31H/N31H_r as indicated (Table 2).Mutations were confirmed by restriction digestion and DNAsequence analysis. The flavin composition of the purified,mutated proteins was determined by HPLC, as described pre-viously (47, 50).Homology Modeling—Homology models of the two Asn-31

mutants were created using the Swiss-Model server (51) on thebasis of the Slr1694 crystal structure (23).Fluorotyrosine Labeling—For incorporation of 3-fluoroty-

rosine (AstaTech, Inc.) and 2-fluorotyrosine (Matrix Scien-tific), a fed-batch fermentation protocol was established.CpX�Y cells transformedwith pET28a(�)-slr1694were grown

TABLE 1Oligonucleotides used for in frame deletion of tyrARegions homologous to tyrA are shown in boldface type.

Primer Sequence (5�3 3�)

DtyrA-5� GGATCTGAACGGGCAGCTGACGGCTCGCGTGGCTTAAGAGGTTTATTATGGTGTAGGCTGGAGCTGCTTC

DtyrA-3� GATGATGTGAATCATCCGGCACTGGATTATTACTGGCGATTGTCATTCGCCGGCTGACATGGGAATTAGC

Redox Modulation of Photoinduced Electron Transfer in BLUF Domains

31726 JOURNAL OF BIOLOGICAL CHEMISTRY VOLUME 287 • NUMBER 38 • SEPTEMBER 14, 2012


ww

.jbc.org/D

ownloaded from

http://www.jbc.org/

in M9 minimal medium supplemented with 50 �M riboflavinand unfluorinated L-tyrosine in a 500-ml fermentation vessel(Multifors reactor, Infors HT AG, Basel) at 37 °C with percola-tion of pressurized air and gentle agitation. By monitoring thepO2 level of themedium via an oxygen electrode, themetabolicactivity of the cells was observed indirectly. Upon consumptionof essential nutrients (glucose, ammonia, tyrosine) the pO2level rises abruptly because oxidative phosphorylation comes toa halt, and the cells start to enter a stationary phase. This eventwas used to lower the temperature to 26 °C and to start feedingnew carbon and nitrogen sources as well as the fluorotyrosineanalogs (160 mg/liter). The protein production was induced bythe addition of 1 mM isopropyl 1-thio-�-D-galactopyranosideshortly after. The cells were further cultivated for about 13 hunder these conditions and harvested subsequently. Proteinpreparation from these cells was carried out as describedpreviously.Spectroscopy—Steady state spectra were recorded on a two-

beam scanning UV-visible photometer (Cary300bio, Varian).Dark state recovery was measured at 493 nm after blue lightillumination using an LED (Luxeon Lumiled, 450 nm, 1 watt).Ultrafast Transient Absorption Spectroscopy and Data

Analysis—Visible absorption spectroscopy was carried outusing pump-probe setups as described previously (44, 52). Thereaction was induced at 400 nm with an energy of �800nJ/pulse. In order to prevent multiple excitations of the samemolecule, the sample was put between two windows separatedby a 200-�m spacer and moved perpendicularly to the probebeam in a Lissajous motion as described previously (14). Beforeanalysis, a preprocessing method was applied to data sets tocorrect for the pre-time 0 signal by subtraction of the averagepre-time 0 signal at each wavelength. The time-resolved datacan be described in terms of a parametric model in which someparameters, such as those descriptive of the instrumentresponse function, are wavelength-dependent, whereas others,such as the lifetime of a certain spectrally distinct component,underlie the data at all wavelengths. The presence of parame-ters that underlie the data at all wavelengths allow the applica-tion of global analysis techniques (53), which model wave-length-invariant parameters as a function of all available data.The partitioned variable projection algorithm is well suited tothe optimization of model parameters for global analysis mod-els (54). The algorithm has the further advantage of estimatingthe S.E. of parameter estimates, an advantage that is useful inmodel selection and validation. A compartmental model wasused to describe the evolution of the spectrally distinct compo-nents in time. Global analysis was then applied to estimate thelifetime and relative concentration of each component at each

wavelength in the data. All data analyses were carried out usingTIMP (54) and the Glotaran software package (55).

RESULTS

Positively Charged Amino Acids Close to C2�O IncreaseCofactor Selectivity in Slr1694—Mutations N31H and N31Rclose to the C2�O carbonyl group of the isoalloxazine ring ofthe flavin were successfully introduced into the BLUF domainof Slr1694. Both mutated proteins show wild type-like groundstate absorption in the dark-adapted state, slightly shifted by 3(N31R) and 4 nm (N31H) (50). The dark recovery is sloweddown by a factor of about 4 in both mutants to about 28 s(N31H) and 29 s (N31R). Despite their similarity to AppA(N31H) and BlrB (N31R), the dark recovery seems not to beinfluenced strongly by these mutations alone. In the AppA-likeN31Hmutant, a drastic slowing down of the dark recovery wasexpected, whereas in the BlrB-like N31R mutant, a behaviorsimilar to that of Slr1694 WT was expected. Interestingly, theintroduction of a positive charge at this position led to a pref-erential binding of FMN and FAD in the BLUF domain of bothAsn-31 mutants (50); BLUF domains heterologously expressedin Escherichia coli usually show a quite heterogenous flavincomposition with similar amounts of riboflavin, FMN, andFAD bound to the photoreceptor domain (47, 56). In the N31Rand N31H mutants, the amount of riboflavin was below thedetection level. From the homology models of the two mutantproteins, the changed residues are within hydrogen bondingdistance of theC2�Ocarbonyl of the flavin (Fig. 1C). Addition-ally, the groups are close to the negatively charged phosphategroup of the flavin, thus supporting its coordination by Arg-30,a semiconserved residue in the BLUF family, which is present inneither BlrB nor AppA.Fluorotyrosine Labeling of Slr1694—2- or 3-fluoro-L-tyrosine

was exclusively incorporated into the BLUF domain using acustom-made tyrosine auxotrophic strain to prevent biosyn-thesis of unfluorinated tyrosine, which would compete mostlikely successfully with the tyrosine analog. In combinationwith a fed-batch procedure, high cell density was achievedalong with complete consumption of the supplemented L-ty-rosine before fluorinated tyrosine was added and the proteinproductionwas induced. After inductionwith isopropyl 1-thio-�-D-galactopyranoside, the cells, if supplemented with naturaltyrosine, usually double about once during the following pro-tein production phase (not shown). However, after supplemen-tation with fluorotyrosine, the cell density increased only byabout 50%, indicating some interference with cell metabolism.The purified 2- and 3-fluorotyrosine-labeled BLUF domainsshowed WT-like dark state absorbance spectra (Fig. 2A). Thedark recovery after blue light excitation was slowed down by afactor of about 4 in 2-fluorotyrosine-labeled Slr1694 (SlrY2F) toabout 40 s, whereas 3-fluorotyrosine-labeled Slr1694 (SlrY3F)was only slightly affected (Fig. 2B).Ultrafast Dynamics of Mutant and Modified Slr1694 BLUF

Domains—TheN31R, N31H, SlrY2F, and SlrY3F proteins wereinvestigated in H2O buffer by transient absorption spectros-copy using 400-nm excitation and white light probe beams.Because the dark recovery reaction is also hydrogen/deuteriumisotope-dependent and slowed down in D2O, we were only able

TABLE 2Oligonucleotides used for site-directed mutagenesis on Slr1694Mismatched base pairs corresponding to the changed amino acid codons are shownin boldface type.

Primer Sequence (5�3 3�)

N31R CTTAGAATCTTCCCAAAGACGTAATCCGGCCAATGGCN31R_r GCCATTGGCCGGATTACGTCTTTGGGAAGATTCTAAGN31H CTTAGAATCTTCCCAAAGACATAATCCGGCCAATGGCN31H_r GCCATTGGCCGGATTATGTCTTTGGGAAGATTCTAAG




ww

.jbc.org/D

ownloaded from

http://www.jbc.org/

to measure SlrY3F in D2O buffer, which has a sufficiently fastdark recovery under these conditions. Because of theirdecreased dark recovery rate (see above), the remaining sam-ples do not allow for a complete recovery in the describedexperimental setting, whichwould lead to amixing of dark- andlight-adapted states. Because of their spectral similarity, bothstates would then be excited by the pump beam, and a mix ofspectral dynamics would be observed.

The main interest of this study is the first step in photoacti-vation of BLUFdomains, which is photoinduced electron trans-fer. This process can be conveniently monitored at around 700nm (Fig. 3). At this wavelength, only the excited state of flavinabsorbs significantly without contributions of anionic or neu-tral flavin semiquinones that are expected to form as observedpreviously (13, 36, 44). A loss of absorbance at this wavelengththerefore corresponds to excited state deactivation processes,

FIGURE 1. Amino acid composition in selected BLUF domains. A, sequence alignment of the first 35 amino acids in the N-terminal part of various BLUFdomains. The conserved tyrosine and the mutated position 31 are shown in boldface type. From the dark-adapted state structure of the WT (B), the predom-inant interactions of the flavin cofactor with the protein are hydrogen bonds from Gln-50 to N5, a hydrogen bond from Asn-31 to C2�O, and hydrogen bondsbetween N3, C4�O, and Asn-32. Homology models (C) of the N31R (orange) and N31H (green) mutants show the putative positive charge of these mutated sidechains near the C2�O carbonyl group. Furthermore, the residues are interacting with the negatively charged phosphate group of the flavin cofactor,supporting its coordination by Arg-30.

FIGURE 2. Absorption and dark state recovery of SlrY2F (gray) and SlrY3F (black). The dark-adapted spectra of the fluorotyrosine-substituted Slr1694 BLUFdomains (A) are highly similar with absorption maxima of the S0-S1 transition at 441 nm identical to the WT. Minor differences are visible due to slight scatteringof the SlrY3F protein indicated by an apparent increase in absorption toward shorter wavelengths. Dark recovery at room temperature after illumination at 450nm was monitored at 493 nm (B). The time constants of the process are about 9 s for SlrY3F and 40 s for SlrY2F.




ww

.jbc.org/D

ownloaded from

http://www.jbc.org/

which are dominated by photoinduced electron transfer inBLUF domains. In Fig. 3, the corresponding traces are depictedalong with the absorbance change of the WT protein at thiswavelength (adapted from data presented by Gauden et al.(13)). All proteins showa clearly elongated excited state lifetimeof the flavin. The excited state decay of all mutants and Slr-Y3Fis similar to the wild type within the first 3 ps, for the Asn-31mutants even within the first 6 ps. Afterward, the decay clearlydeviates from the wild type. Among these proteins, SlrY3F inboth D2O and H2O shows the longest excited state lifetime.SlrY2F showed very similar photodynamics to SlrY3F on theultrafast time scale (not shown) and is therefore not discussedhere.Spectral Evolution—To obtain an overall picture of the spec-

tral evolution and to assess whether reaction intermediates canbe observed in the obtained data sets, we first analyzed the dataglobally using a sequentialmodel with increasing lifetimes (1323 33 4 . . . ). The corresponding evolution-associated dif-ference spectra (EADS) are displayed in Fig. 4. The followingapplies to all data sets. The first spectra (black) correspondpurely to the singlet excited state of the flavin that is formedduring the instrument response. The main components ofthese spectra typically are the ground state bleach (GSB)around 445 nm, excited state absorption (ESA) around 510 nmand above 600 nm, and stimulated emission (SE) at around 550nm. The final, non-decaying spectra correspond to a speciesthat does not decay on the time scale of the experiment and ispredominantly assigned to the difference spectrum of the sig-naling state of the BLUF domain, as indicated by the dark statebleach around 445 nmand the red-shifted absorption at around490 nm. In all cases, a featureless absorption from 500 to 700nm is observed in varying amounts, which is attributed to flavintriplet absorption (37). The spectral evolution from the black tothe final spectra will be described for all studied BLUF domainsas follows.

The Slr1694-N31R and N31H mutants behave similarly intheir spectral evolution and are both sufficiently describedusing five lifetimes (Fig. 4, A and B). The red EADS is formedfrom the black in roughly 1 ps (N31H) and 1.2 ps (N31R) alongwith a significant blue shift and increase of the SE band fromabout 560 to 550 nm, which is indicative of vibrational relax-ation in the flavin excited state (13, 36, 37, 44). The red EADSthen decays in 5.1 ps (N31H) and 5.5 ps (N31R) into the greenEADS with a loss and further blue shift of the SE band. Addi-tionally, the green EADS gain absorption between 570 and 620nm, which indicates the formation of a flavin semiquinone spe-cies, as observed previously (13, 44). Its rise and decay are indi-cated at single wavelength traces around 600 nm (supplementalFig. S1). At the same time, ESA above 620 nm is diminished byabout 25%. From the green EADS, the blue EADS is formed in36 ps (N31H) and 37 ps (N31R), which is characterized by a lossof GSB of about 50% and a complete (N31H) or almost com-plete (N31R) loss of SE. Additionally, a shoulder in the positiveabsorption at around 490 nm is formed, along with a furtherincrease in absorption between 550 and 600 nm. The finalmagenta EADS is formed in about 240 ps in both mutants witha quantum yield of about 32% (N31H) and 29% (N31R), asjudged from the induced absorption at 490 nm relative to theGSB in the very first spectrum compared with previous exper-iments (44). The N31H data set seems to contain about 2 timesmore flavin triplet species than N31R, as judged by the broadabsorption near 650 nm.The spectral evolution of SlrY3F (Fig. 4,C andD) and SlrY2F

(not shown) is highly similar. We will therefore focus only onthe description of the SlrY3F sample. Similar to the spectra ofthe Asn-31 mutants, the red EADS is formed from the black in1 ps with an increase and a slight blue shift of SE. The red EADSthen evolves into the greenwith 2.7 and 4.3 ps (D2O). The greenEADS shows a significant blue shift of the SE band. Addition-ally, the ESA feature at around 510 nm becomes narrower and

FIGURE 3. Excited state decay of the flavin in Slr1694 and mutants. The absorbance change at 701 nm is characteristic for excited state decay of the flavin.Slr1694-N31R/N31H (A, black) show a significantly slower decay than the WT (gray), especially at delays greater than 10 ps. SlrY3F in both H2O and D2O showsan even slower excited state decay (B).




ww

.jbc.org/D

ownloaded from



http://www.jbc.org/

shows a shoulder at around 490 nm. Similar to the spectralevolution of the N31R and N31H mutants, some absorptionbetween 580 and 600 nm rises (see also supplemental Fig. S2),but to a significantly smaller degree,whereas ESAabove 625nmremains unchanged. The green EADS evolves into the bluespectrumwith a lifetime of 31 and 55 ps (D2O)with a�50% lossof GSB and ESA. SE is also diminished but still clearly present.The blue EADS then evolves in 365 and 689 ps (D2O) into thenon-decaying species. Compared with the data set of the N31Rmutant, the final spectrum of the SlrY3F data sets shows alarger amount of featureless absorption between 500 and 700nm. The apparent broadening of the induced absorption at 490nm, which is indicative of the BLUF signaling state, is due tospectral overlap with triplet absorption. As judged from theabsorption at 650 nm, the SlrY3F protein yields similaramounts of triplet in H2O and D2O, similar to the N31Hmutant. The quantum yield of signaling state formation esti-mated as above is significantly lower than for the Asn-31mutants with about 16%.Target Analysis of the N31R andN31HMutants—In the past,

target analysis proved to be a powerful tool to reveal the ultra-fast photochemistry of BLUF domains (13–15, 36, 37, 43, 44).By fitting transient data using branched compartmental mod-els, parallel reaction pathways can be investigated, and invertedkinetics, where a product is decaying faster than its formation,can be addressed. The resulting so-called species-associateddifference spectra (SADS) ideally represent the difference spec-tra of the truemolecular species that occur during the reaction.Using this approach on transient absorption data of Slr1694,the presence of an anionic and a neutral semiquinone flavinradical was demonstrated previously (13, 14). Due to the strongmultiexponentiality of the excited state decay in BLUF domainsranging from a few ps to hundreds of ps, the electron transfer

from a nearby tyrosine and the resulting anionic flavin radicalproduct, which occur in a few ps, are difficult to observedirectly. Additionally, the subsequent protonation of the ani-onic flavin radical also occurs in only a few ps. Because also theprotonated semiquinone lives only for tens of ps, either inter-mediate is difficult to resolve kinetically. A multiexponentialexcited state decay, as previously observed for BLUF domains(13, 14, 36, 44), is clearly visible in the transient absorption at701 nm (Fig. 3). To obtain spectra of the pure intermediatestates, we applied target analysis taking this behavior intoaccount. Additionally, we considered vibrationally hot excitedstate relaxation because it was indicated by the blue shift ofstimulated emission observed in the sequential analysis above.Finally, a non-decaying species is assumed for the red-shiftedsignaling state BLUFred, which however may be mixed with tri-plet features that are not expected to decay on the time scale ofthe experiment and are therefore impossible to separate fromeach other. In the model for the N31H mutants (Fig. 5A), twointermediates (Q1 and Q2) formed sequentially from the mul-tiexponential excited state decay were included, identical to themodel for theWT (13, 44). SlrY3F, however, was best describedwith only one intermediate (Q1) in bothH2O andD2O (Fig. 5A)and is therefore discussed separately. Similar to WT, we alsoincluded a 50% loss after the Q1 intermediate, ascribed to rad-ical recombination prior to proton transfer (13).To get a better handle for discussion of the obviously slowed

down excited state decay rates, as already observed in the rawdata (Fig. 3), we describe this multiexponential process in thefollowing by averaged lifetimes calculated as follows,

kavg � e�n

ln�kn� � fn (Eq. 1)

where kn represent the respective decay rates, and fn are thecorresponding fractions obtained by target analysis. The life-

FIGURE 4. Spectral evolution of Slr1694-N31R (A), Slr1694-N31H (B), and SlrY3F in H2O (C) and D2O (D). The EADS show the spectral evolution afterfemtosecond excitation. The spectra evolve sequentially with the indicated lifetimes (black3 red3 green3 blue3magenta).




ww

.jbc.org/D

ownloaded from


http://www.jbc.org/

times and their corresponding fractions obtained from the tar-get analysis below are displayed in Table 3. The average life-times (�avg) displayed in Table 3 and Fig. 5A are the reciprocalvalues of kavg. The logarithmicway ofweight averaging the ratesprovides a more unbiased average than weight averaging eitherthe lifetimes or the decay rates directly, because using the life-times will put more weight on the bigger lifetimes, whereasusing the rateswill emphasize the shorter lifetimes/bigger rates.The data sets of the Asn-31 mutants were best fitted with a

model identical to the WT as shown in Fig. 5A and gave quali-tatively identical spectra for bothmutants (Fig. 5B). Thismodelincludes two intermediates (Q1 and Q2). The hot (black) andrelaxed (red) excited state SADS represent flavin excited stateswithGSB around 445 nm, ESA absorption at 510 and above 600nm, and stimulated emission around 550 nm. The relaxed stateis formed in 1 ps in both mutants and characterized by a blue-shifted SE band. The decay of the relaxed excited state into theQ1 intermediate was described best using four lifetimes. Thelifetimes differ only slightly between the two data sets (Fig. 5A).Compared with the wild type (Table 3), the average lifetime isslowed down to 63 and 65 ps (WT, 17 ps) due to lower fractionsof the fast components and the presence of clearly elongatedlifetimes with significant contributions.The Q1 intermediates in the N31H and N31R mutants sig-

nificantly differ in their spectral properties from the Q1 inter-mediate detected inWT. The spectrum is characterized by thesame induced absorption above 550 nm, which is characteristicfor an anionic semiquinone species with charge transfer (CT)character (57). Strikingly, the Q1 SADS shows a negativeband at 525 nm. Negative bands in transient absorptionspectroscopy are assigned either to GSB or SE, and given that

there is no ground state absorption at 525 nm, the negativefeature must be assigned to SE. Thus, we conclude that thisspecies, at least partly, corresponds to a flavin excited state.At first glance, one may interpret the Q1 SADS as a WT Q1spectrum that is “contaminated” by FAD*. However, this canbe ruled out because the negative feature at 525 nm has aspectral shape that differs significantly from that of FAD*.Additionally, the spectral evolution of N31H clearly indi-cates the formation of a shifted SE feature in contrast to WT(supplemental Fig. S3). A target model without this interme-diate led to a significantly worse fit in this time domain (notshown), which proves that inclusion of an emission compo-nent in Q1 is strictly required.The Q1 intermediates decay into the next intermediate (Q2)

with WT-like lifetimes of 4.4 ps (N31H) and 5 ps (N31R). TheQ2 intermediate nicely represents a neutral semiquinone flavinradical, which is characterized by the complete absence of SEand a broad absorbance between 550 and 650 nm. Similar toWT, we observe a significant loss in GSB in both data sets (13,14, 44) if the 50% loss mentioned above is not included. Thisloss is believed to originate from radical recombination of theanionic flavin semiquinone and tyrosyl cation, thus preventingquantitative formation of the neutral semiquinone. The Q2intermediates decay with 54 ps (N31H) and 63 ps (N31R) intothe final non-decaying spectrum. This values are in close rangeof the 65 ps observed for the neutral flavin semiquinone inWT(13). The final non-decaying spectrum is identical to the oneobtained from global analysis and shows the characteristicinduced absorption of the red-shifted signaling state at around490 nm and a broad featureless absorption most likely corre-sponding to a flavin triplet.

FAD*(hot)

FAD* FAD* FAD* FAD*

Q1

Q2

BLUFred

A B

C

1 ps

N31H 4.4 psN31R 5 ps

SlrY3F(H2O) 15 ps(D2O) 28 ps

N31H 54 psN31R 63 ps

SlrY3F(H2O) 88 ps(D2O) 100 ps

N31H 63 psN31R 65 ps

FIGURE 5. Target analysis of Asn-31 mutants of Slr1694 (A and B) and Slr1694 substituted with 3-fluorotyrosine SlrY3F (A and C). The model used for thedescription of both N31H (B, solid lines) and N31R (B, dashed lines) is displayed in A together with the average lifetimes for the multiexponential reaction fromthe FAD* excited state (red) to Q1 (blue). The model used for the description of SlrY3F in both H2O (C, solid lines) and D2O (C, dashed lines) data sets is similar tothe N31H/R model (A) except that the reaction proceeds from Q1 directly to the red-shifted state (BLUFred).




ww

.jbc.org/D

ownloaded from


http://www.jbc.org/

Target Analysis of the SlrY3F Protein—For the fluoroty-rosine-substituted protein, we extracted an intermediate,which is strikingly similar to the Q1 spectrum of the Asn-31mutants, varying slightly by an increased absorbance above 650nm and at 510 nm. The model (Fig. 5A) again features hot state(black) cooling into the relaxed excited state (red) and forma-tion of the non-decaying species (magenta) via the Q1 interme-diate (blue; Fig. 5C) without any further intermediate. Similarto theAsn-31mutants, the excited state spectra perfectlymatchflavin excited state features with a blue shift of the SE bandduring the 1-ps hot state relaxation. The excited state decayinto the Q1 intermediate was described best using four excitedstate lifetimes with varying concentrations (Fig. 5A and Table3). The kinetic isotope effect on the slower components is small(1.1–1.3) and is not further interpreted. Compared with WTand theAsn-31mutants (Table 3), the average lifetime is sloweddown further to 88 ps (100 ps in D2O) due to the dominatingcontributions of the slow components in the nanosecond timescale. The semiquinone characteristics of the Q1 intermediateat around 600 nm are clearly observed in kinetic traces aroundthis wavelength (supplemental Fig. S2) and support the globalfitting procedure. The decay of the Q1 intermediate at 600 nmalso shows a clear kinetic isotope effect of about 1.9, which issupported by the raw data (supplemental Fig. S2). The finalspectra in both data sets are very similar and resemble the red-shifted signaling state together with contributions of rather fea-tureless absorption between 500 and 700 nm assigned to flavintriplets.Spectral Fitting of the Q1 Intermediate—The Q1 spectrum

observed here is clearly distinguished from theQ1 intermediateobserved in theWT (13) andW91F (44)mutant by pronouncedadditional negative features at �510 and 540 nm, which indi-cate a stimulated emission contribution in this species. Thisobservation implies that this SADS represents a fraction of fla-vins in an excited state. To identify the nature of the Q1 inter-mediate, we reconstructed the difference spectrum by spectralfitting of a linear combination of experimentally known contri-butions and spectral characteristics of excited state absorptionand stimulated emission of flavins in BLUF domains. We usedexperimentally determined spectra of the ground state absorb-ance to describe the GSB and skewed Gaussians to describestimulated emission and (excited state) absorbance (Fig. 6). InFig. 6, a comparison of spectral fits of the Q1 intermediate ofN31H, N31R (Fig. 6,A and B), SlrY3F (Fig. 6,C andD), andWT(Fig. 6, E and F), the latter two in H2O and D2O, is displayed. InWT, theQ1 spectrum is sufficiently described using the groundstate bleach (green) and three skewed Gaussians (blue, red, andmagenta) for the absorption of the intermediate in H2O and

D2O. In themodified proteins, an additionalGaussian (black) ataround 525 nm was necessary to account for the negative con-tributions at 500 nm and 540 nm. This additional contributionat �525 nm can only originate from stimulated emission, dem-onstrating thatQ1 corresponds at least partly to an excited statespecies. In supplemental Fig. S4, we present a fit of all photo-cycle intermediates for the N31H mutant, with known contri-butions from ground state bleach, stimulated emission, andred-shifted product absorption imposed in the fit. The FAD*species shows stimulated emission from the excited state to theground state, peaking at �510 nm, as follows from the fluores-cence spectrum,whereasQ2 and the final product are devoid ofany stimulated emission. The SE observed in the Q1 interme-diate is therefore present only in this intermediate and red-shifted compared with the locally excited state FAD*. A redshift in fluorescence/SEmay be indicative for the formation of aCT state.

DISCUSSION

Up to now, only a few BLUF domains were studied withregard to the primary photochemistry. Therefore, the primarymechanisms that facilitate formation of the biological signalingstate and determine most likely also biologically relevantparameters like photosensitivity (quantum yield of signalingstate formation) are still not fully understood. Although theBLUF domains share a very similar amino acid composition oftheir flavin binding pocket, the differences in excited state life-times ranging from about a few ps in Slr1694 (13) to hundredsof ps in AppA (37) have not been addressed on a molecularlevel. In this study, we observed how the redox potential of thetyrosine/flavin reaction pair, which constitutes the first step inBLUF photoactivation, determines excited state lifetime as wellas photoproduct quantum yield.Observation of an Excited State Charge Transfer State in

BLUF Domains—The elongated excited state lifetimes in theAsn-31 mutants and SlrY3F provided us with further insightinto the BLUF photocycle by revealing a hitherto unobservedexcited state. In both Asn-31mutants and SlrY3F, we found thepresence of an intermediate (Q1), which features the previouslyobserved charge transfer absorption of the anionic flavinsemiquinone but additionally shows stimulated emission fea-tures indicative of an excited state species. By spectral fitting ofthe Q1 intermediates observed here and the Q1 intermediatesobserved in WT before, we clearly assign the difference tostimulated emission that is red-shifted to �525 nm com-pared with that of the locally excited state FAD*. Therefore,we believe that the Q1 intermediate indicates the presence ofan excited state with strong charge transfer character, which

TABLE 3Excited state decay lifetimes (�) of various BLUF domains, their fractional contributions, and the corresponding weighted average lifetimes(�avg)The last row shows the quantum yield of signaling state formation (red).

Slr1694WT (13) Slr1694-N31H Slr1694-N31R SlrY3F SlrY3F (D2O) AppA (37) BlrB (36)

� (ps) 7 (47%) 6.7 (20.6%) 6.9 (20.6%) 2.8 (7.3%) 2.8 (7.3%) 90 (38%) 18 (27%)40 (28%) 37 (34.6%) 43 (34.6%) 11.5 (26.6%) 11.5 (26.6%) 570 (59%) 216 (73%)180 (17%) 179 (38.2%) 181 (38.2%) 101 (34.5%) 111 (34.5%) 1000 (3%)209 (0.1%) 3448 (6.6%) 1667 (6.6%) 943 (31.5%) 1266 (31.5%)

�avg (ps) 17 63 65 88 100 287 110red 40% 33% 29% 16.5% 15% 24% 30–40%




ww

.jbc.org/D

ownloaded from




http://www.jbc.org/

is formed from the locally excited state (FAD*) prior to fullelectron transfer (Fig. 7A). Interestingly, fluorescence depolar-ization experiments in glutathione reductase provided evidencefor an emissive flavin/tyrosine CT state with a shifted emissiondipolemoment decaying in the same timedomain (58).Hence,wehypothesize that the excited CT state observed here most likelycorresponds to significant electron redistribution from Tyr-8 toFAD in the excited state.The Q1 intermediate in the Asn-31 mutants has a spectral

signature of the anionic FAD. CT absorption band near 600 nmthat is very similar to that ofWT.Moreover, its lifetime of�5 psis essentially the same as in WT. Therefore, it is likely that theQ1 species is not a single molecular species but that it repre-sents amixture of an excitedCT state with the anionic semiqui-none FAD. . Such a mixture may appear in the case of hetero-geneity in the initial reaction rates, which in a fraction of theBLUF domains results in transient accumulation of the excitedCT state and in the remaining fraction in transient accumula-tion of the FAD. species.

The question arises whether the proposed FAD*/Tyr excitedCT state forms an integral part of the photoreaction in WTBLUF or if this species is a characteristic feature of themodifiedBLUF proteins only. Given that functionally the photoreactionsinN31R andN31Hare the same as inWT, it is very likely that inWT, the reaction proceeds via such a species as well but that itcannot be kinetically resolved due to a rate-limiting formationrate.We therefore propose that the FAD*/Tyr excited CT stategenerally applies to the BLUF photoreaction and that itbecomes observable in the N31R/H mutants and Slr-Y3F pro-tein through their modified reaction rates.The Q1 species is subsequently protonated to the neutral

semiquinone. In case of the Asn-31 mutants, one can safelyassume that the proton transfer reaction from the tyrosineremains largely unaffected, which is indicated by the virtuallyidentical rise time and lifetime of the neutral flavin semiqui-none (Q2) in the WT.The Q1 spectrum of SlrY3F is highly similar to the Q1 inter-

mediate in the Asn-31 mutants and therefore considered to be

FIGURE 6. Spectral fits of the Q1 intermediates of the Asn-31 mutants (A and B) and SlrY3F (C and D) compared with WT (E and F) using experimentallydetermined ground state absorbance (green) and skewed Gaussians for (excited state) absorption (blue, red, and magenta) and stimulated emission(black).




ww

.jbc.org/D

ownloaded from

http://www.jbc.org/

of almost identical nature. The decay of this species is sloweddown 3-fold compared with WT Q1 and shows a clear kineticisotope effect of 1.9. Due to the higher acidity of the hydroxylgroup of the fluorinated tyrosine (59), one would assume thatthe proton transfer to the flavin is even faster than in WT andthe value obtained here by global fitting. This would then leadto an accumulation of the neutral radical form, which is appar-ently not the case. Thereby, the subsequent radical recombina-tion step, which involves proton-coupled electron transfer toform the red-shifted state apparently occurs in the same timedomain as its formation. Compared with WT, this reaction isaccelerated here, which is reasonable because the fluorinationrenders the tyrosine radical in an energetically less favorableredox state. Small contributions from a neutral semiquinonemixed with the CT and anionic semiquinone spectrum mightaccount for the increased absorbance at 510 nm in SlrY3F Q1compared with N31H/R-Q1 (Fig. 6), which was previouslyobserved to be higher in the neutral semiquinone than in theanionic form (13).With these new experimental insights at hand, we propose a

more detailed photocycle scheme for BLUF domains (Fig. 7A).After excitation of the flavin, significant charge redistributionfrom tyrosine to flavin takes place before the complete electrontransfer is accomplished. In Slr1694 WT, electron transfer istoo fast to kinetically resolve this excited state FAD*/Tyr CTintermediate and seems to directly proceed to the ground stateanionic flavin semiquinone. In Slr1694 proteins with sloweddown electron transfer properties, this intermediate becomespartly detectable in transient absorption spectroscopy by itsred-shifted SE contribution. Because of its otherwise spectraland kinetic similarity to the WT flavin anionic semiquinonespecies, it is most likely that we observe a mixture of excitedFAD*/Tyr CT state and flavin anionic semiquinone/tyrosylradical pair.Positive Charges Close to the Flavin C2 Carbonyl Slow Down

Electron Transfer in Slr1694—Besides distance, orientation,and reorganization energy, the relative redox potentials in a

given reaction partner system are of major influence for thefree reaction energy and thereby also for electron transferrates. The redox potential of a molecule can be influencedindirectly by changes in the environment (e.g. polarity anddielectric constant) or by subtle chemical modifications ofthe molecule itself. In a biological context, redox processesare evolutionarily optimized and present the best fittingenvironment for the corresponding task. The redox poten-tial of flavin is known to be strongly influenced by positivecharges close to the C2 carbonyl of the flavin, which has beenobserved previously in many flavoenzymes (60, 61). The pos-itive charge close to this part of the flavin is thought to sta-bilize negatively charged flavins because the charge densityin flavins is generally more localized on the heteroatom-richpart of the isoalloxazine moiety, especially in the excitedstate or in charge transfer states. So far, redox potentials offlavin in BLUF domains have been experimentally addressedonly for AppA and several mutants thereof (62). The redoxpotential for the ground state of the FAD/FAD redox pairin the WT BLUF domain was determined to be about 260mV (versus SCE), about 50 mV lower than for FAD in solu-tion. Interestingly, BLUF domains are highly variable in theiramino acid composition close to the C2 carbonyl (Fig. 1A).Slr1694, which so far shows the fastest excited state decay(Table 3) and therefore also the fastest photoinduced elec-tron transfer, contains a polar but uncharged asparagine atposition 31 (Fig. 1, A and B). BlrB and AppA, which show aslightly and significantly slower photoinduced electrontransfer, respectively, contain an arginine or a histidine atthe corresponding position (36, 37). Both side chains aresupposed to be positively charged in the protein under theexperimental conditions applied here. The protonation stateof His-44 in AppA was previously confirmed by NMR spec-troscopy (63). Assuming that these positive charges stabilizealso a singly reduced flavin, one would expect an increase inthe FAD/FAD. redox potential and thereby a correspondingincrease in free reaction energy. In contrast, a significantly

A B

e -

Y8

FAD

N31 N31R

N31H

SlrY3F

O

NH

N

O

R

N

N

H 3 C

H 3 C

3

F OH

H 2 N O NH 2

+

HN H 2 N

HN

N + H

τavg (Φred)

WT 17 ps (40%)N31R 63 ps (33%)N31H 65 ps (29%)SlrY3F 100 ps (16%)

FAD

FAD*

FADH•

FAD•–

FAD*(CT)

BLUFred

WT 60 psN31H 58 psN31R 63 psSlrY3F n.d.

WT 6 ps (18)

N31H 4 psN31R 5 psSlrY3F 15 ps (28)

WTs

FIGURE 7. A, in the Slr1694 Asn-31 mutants and modified proteins, the slowed down excited state decay leads to an accumulation of a charge transferstate FAD*CT formed from the locally excited state FAD* prior to formation of the anionic semiquinone (FAD.), which appears to be formed instantly inWT. The lifetimes in the modified proteins correspond to a mixture of CT and anionic radical species. In the Asn-31 mutants, the subsequently formedneutral flavin semiquinone (FADH�) decays to the red-shifted state with similar lifetimes as the WT. In SlrY3F, this intermediate cannot be observed, andthe red-shifted state appears to be formed directly from the FAD*CT/FAD. . The average excited state lifetime (�avg), which corresponds to excited stateelectron transfer between tyrosine and flavin in BLUF domains (B), is slowed down by modification of the electron donor (SlrY3F) or a positive chargenear the C2�O carbonyl of the acceptor (N31R and N31H). Additionally, the quantum yield of signaling state formation (red) becomes lower withincreasing excited state lifetime.




ww

.jbc.org/D

ownloaded from

http://www.jbc.org/

slowed down photoinduced electron transfer is observed forthese WT proteins and the Slr1694-N31R/H mutants (Fig.7B and Table 3). If we consider the overall protein configu-rations and especially the mutual distance and orientation offlavin and tyrosine identical to that in the WT protein, thisputatively redox potential-induced change in reactivity maybe explained by the Marcus theory (64–66). The increase infree energy by elevating the redox potential of the flavinmight push the reaction into the so-called inverted region. Inthis region, an activation barrier appears, which then slowsdown the reaction itself but still yields a higher free energy.Generally, the reorganization energy for complex systems likeproteins is very high, and therefore it is usually hard to reach theinverted regime. Here, however, we are looking at an excitedstate reaction, which most likely provides sufficient energy; theredox potential in the singlet excited state of flavin is elevatedenormously by �1 V (67). Additionally, Marcus invertedregions have also been found in ET reactions of photosyntheticreaction centers (68) and have been recently observed for theback electron transfer reactions to the neutral flavin semiqui-none in DNA photolyase after DNA repair as well (69).It should be noted that although we observe a significant

effect on the ET rates, the redox potential of the one-electronreduction of flavin might be shifted only slightly. In a previoustheoretical study on flavodoxins, a shift of 20mVwas calculatedfor mutants, where charged or polar side chains were intro-duced close to the flavin (45).Theaverageexcitedstate lifetimesofN31H/Rwith63and65ps,

respectively, are significantly slower than in WT (17 ps) but stillclearly faster than BlrBwith 110 ps or AppAwith 287 ps. Interest-ingly, mutated AppA-H44R, which turns AppA into a BlrB-likeprotein with respect to the C2�O environment, leads to a signifi-cant speeding up of the primary reaction (39). In Slr1694, we areobviously not able to discriminate the AppA-like mutation N31Hand the BlrB-like mutation N31R because the difference in theexcited state lifetimes of all fractions is only marginal (Table 3).Redox Modulation of the Electron-donating Tyrosine in

Slr1694—In tuning the redox properties of the BLUF domain,we also applied a different approach directly affecting the tyro-sine redox partner, which is a well known procedure for modi-fying and investigating PCET in photosynthesis research (33).In this study, we used a procedure in which a tyrosine analog,fluorotyrosine, is incorporated selectively into the protein invivo using a tyrosine biosynthesis-deficient E. coli expressionstrain. A similar procedure was previously employed to modifythe GFP chromophore (70). Here we decided to design a noveltyrosine-deficient E. coli expression strain using CmpX13 (47),a C41(DE3) (71) derivative with constitutive expression of ariboflavin transporter, as a parent strain. This strategy is advan-tageous for the production of flavoproteins under the condi-tions encountered here. For selective labeling, the cells haveto be cultivated in defined/minimal medium, which is usu-ally not optimal for cell vitality. Additionally fluorinatedtyrosine, which is present during protein production, is alsoincorporated into household proteins of the cell and mightthereby impair their function. Both conditionsmight also leadto a lack or even a breakdown of cofactor biosynthesis, which iseasily circumvented by external supplementation of riboflavin

in themedium. In our experiments, we indeed observed a lowervitality of the expression culture, indicated by the lowered bio-mass yield.The introduction of a fluorine atom at the tyrosine ring

resulted in a functional, photoactivatable BLUF protein. Thischemical modification not only increases the redox potentialfrom 650 to 700 mV (versus NHE) of the Tyr�/Tyr redox pairbut also lowers the pKa of the phenolic group by more than 1order of magnitude to about 8.4, thus increasing its acidity (59).Additionally, one should take into account that although thefluorine atom is very compact and the substitution can be con-sidered by and large isosteric, the fluorine substitution inversesthe polarity of the former C–H bond andmay therefore also actas a hydrogen bond acceptor and establish new interactionswithin the protein (72). Because the protein still binds flavinsimilar to the WT protein and the dark-adapted state absorb-ance and also the dark recovery in at least the SlrY3F proteinremains unchanged, one can assume that the flavin bindingpocket is not significantly distorted by this tyrosine analog. Theincreased redox potential of the fluorinated tyrosine stabilizesits reduced form and is thus expected to transfer its electron tothe excited state flavin more slowly, as indeed is observed here.It is also noteworthy that in AppA, the redox potential of theflavin is largely unaffected by the substitution of tyrosine byphenylalanine, which suggests that we can safely assume thatonly the redox potential of the tyrosine and not the redoxpotential of the flavin is changed by this modification (62). Theaverage excited state lifetime (Table 3) of 88 ps (100 ps in D2O)is clearly longer than for the Asn-31 mutants and close to theBlrB average lifetime (110 ps/120 ps in D2O (36)). The effect ofchanges in the redox potential of the tyrosine on the photoin-duced electron transfer therefore appears larger than the indi-rect effect of the environment of the flavinC2�Ocarbonyl. Theenvironment of the reactive tyrosine in BLUF domains, how-ever, has not been studied so far. From our experience,mutagenesis close to this tyrosine is generally difficult toaccomplish without disturbing the protein fold.5

Studies on the environment of Tyr-8 would be highly inter-esting because another influence for the redox potential ofTyr-8 may be the nature of hydrogen bonding to the conservedGln-50 side chain amide. In a theoretical study by Ishikita (46)on the ET driving force in light- and dark-adapted states of theAppABLUFdomain, the redox potential of the tyrosine seemedto be influenced significantly by the side chain orientation ofthe conserved glutamine, whereas the flavin remained largelyunaffected. The ET driving force would thereby be significantlyenhanced for the light state, which is consistent with the resultsof Toh et al. (73), who observed a dramatically increased elec-tron transfer rate in the light state. It is difficult, however, todirectly relate the results onAppA to the Slr1694 BLUF domainbecause photoinduced ET rates are much faster in the latterprotein (17 ps versus 287 ps; see Table 3). In fact, in TePixD,which is highly homologous to Slr1694, ET in the dark-adaptedstatewas found to occur at optimal�G on the top of theMarcuscurve (42). The ET rate of the dark-adapted AppA is signifi-

5 T. Mathes, I. H. M. van Stokkum, M. Stierl, and J. T. M. Kennis, unpublishedobservation.




ww

.jbc.org/D

ownloaded from

http://www.jbc.org/

cantly slower and may be far off the maximum of the Marcuscurve accordingly. Therefore, the enhanced ET rate in Slr1694in the light state (which occurs very rapidly in 1 ps (15)) is mostlikely due to other factors like shortened Tyr/FAD distance oroptimized MO overlap.Because the change of the redox potential of the tyrosine and

flavin leads to a lower or higher free reaction energy, respec-tively, and ET is slowed down in both cases, ET inWT Slr1694must be optimized almost perfectly and occur in an almostbarrier-less manner according to the Marcus theory. A similarproposal has been made by Shibata et al. (42), who studied thehighlyhomologousTePixDBLUFproteinbytime-resolvedfluo-rescence spectroscopy.Previously, we modulated the tyrosine redox partner by

exchanging the phenolic side chain by an indole moiety usingsite-directed mutagenesis (44, 74). Although the Slr1694-Y8Wmutantwas able to photoreduce the flavin to a radical statewithhigh efficiency, no signaling state was formed, probably due toalterations in the flavin-coordinating hydrogen bond networkinvolving Gln-50. The light-induced reaction produced variousspectrally distinct radical pair difference spectra on the ultrafasttime scale, which were assigned to FAD/Trp-8 radical pairs butmight also correspond to a radical pair consisting of FAD andthe semiconservedTrp-91.Using transient EPR spectroscopy, astrongly coupled radical pairwas detected in thismutant, whichwas significantly different from the WT radical pair anddecayed slightly slower (74). Because the Y8Wmutation, how-ever, rendered the BLUF photoreceptor non-functional and theradical pairs probably originate from the FAD triplet staterather than the singlet excited state, these findings are difficultto relate to WT and functional BLUF mutants directly.The Quantum Yield of Signaling State Formation Is Affected

by Redox Modulation of FAD/Tyr—An interesting aspect ofthis study is the obvious decrease of the quantum yield of sig-naling state formation alongwith the elongation of excited statelifetime (Table 3 and Fig. 7B). In Slr1694, which shows the fast-est excited state decay, we observe a quantum yield of roughly40%, and in BlrB the quantum yield ranges between 30 and 40%(36, 75) as compared with about 24% AppA (37), which has theslowest ET reaction observed in BLUFdomains so far. InAppA,the quantum yield can be increased to about 30% by introduc-ing the BlrB-likeH44Rmutation (39), analogous to the Slr1694-N31R mutation. Additionally, an increase up to 37% can beachieved by removal of a semiconserved tryptophan side chainthat competes in the excited state electron transfer with thetyrosine but only yields a futile reaction that does not contrib-ute to signaling state formation (43). Because this competingprocess was not observed in Slr1694, which is probably due to alarger distance of this tryptophan from the flavin as comparedwith AppA (44), we also do not consider it in the Slr1694 vari-ants investigated here. However, due to the significantly elon-gated lifetime of the flavin excited state, we cannot exclude com-peting ET from Trp-91 completely. The quantum yield in themodified Slr1694proteins drops as a result of changes in the redoxpotential from 40% in the WT to �30% in both Asn-31 mutantswithclearly elongatedaverageexcited state lifetimesof about65psdown to �16% in the fluorotyrosine-substituted protein with aneven longer average excited state lifetime of about 100 ps.

This correlation is also of interest for the question regardingwhich step in the photocycle promotes the putative rotation oftheGln-50 side chain amide and therefore constitutes a prereq-uisite for the signaling state. Ishikita (46) reported calculationson the AppA BLUF domain in putative light and dark stateconformations, which suggested that the light-induced chargeseparation forms the main driving force for the rotation. Gen-erally, this would be in line with our finding here that adecreased ET rate correlates with a lowered signaling statequantum yield. However, the lowered signaling state quantumyield might well be due to various loss processes in the photo-cycle like an overall reduced quantum yield of the light-inducedET reaction or an enhanced radical pair recombination of Y.�/FAD. . Both would contribute to a loss in the subsequentproton transfer to form FADH�, which we previously suggestedto be a trigger for Gln-50 side chain rotation due to disruptionof the hydrogen bond from Gln-50 to FAD N5 (13, 14).The correlation between primary photochemical properties

and quantum yield of signaling state formation supports thenotion that the various BLUF photoreceptors have been evolu-tionarily adapted for their specific physiological function. Espe-cially, the region around the C2�O carbonyl seems to be a keyplayer for reactivity tuning. So far, the physiologically best-studied BLUF containing protein is AppA, which integratesboth light and redox stimuli (5, 6, 76–81). The latter are mostlikely perceived by a C-terminal cysteine-rich domain and/or arecently discovered novel heme binding domain (82), which islocalized between the BLUF domain and the PpsR interactiondomain. Both domains are susceptible to redox changes in theenvironment and are able to induce structural transformationsaccordingly. The extent to which the redox potential of theenvironment directly affects the BLUF domain has not beenaddressed yet. As we observed here, the redox potential of bothflavin and tyrosine determines the quantum yield in BLUF pho-toactivation. If these redox potentials are indirectly coupled tothe environment, onemay include another redox input into theintegrated signal, which originates directly from the BLUFdomain. Such a redox relay, however, has not been observedexperimentally so far.In conclusion, we demonstrate that the redox potential of the

flavin/tyrosine redox pair in BLUF domains is a key determi-nant of excited state electron transfer. By modulation of theredox potential, we provide experimental evidence for a previ-ously unobserved excited state charge transfer intermediateprior to electron transfer in theBLUFphotocycle. Furthermore,the electron transfer rate correlates with the quantum yield ofsignaling state formation. Therefore, the redox potential of thetyrosine/flavin redox pair is directly coupled to the biologicaloutput. The biological reason for this divergent behavior inexcited state decay in various wild type BLUF domains, how-ever, needs to be addressed in future studies.

Acknowledgments—We thank Roman Fudim (Humboldt University,Berlin) for assistance during the measurements and Peter Hegemann(HumboldtUniversity, Berlin) for providing access to his biochemistryfacility and general support.




ww

.jbc.org/D

ownloaded from

http://www.jbc.org/

REFERENCES1. Gomelsky, M., and Klug, G. (2002) BLUF. A novel FAD-binding domain

involved in sensory transduction inmicroorganisms.Trends Biochem. Sci.27, 497–500

2. Gomelsky, M., and Hoff, W. D. (2011) Light helps bacteria make impor-tant lifestyle decisions. Trends Microbiol. 19, 441–448

3. Fiedler, B., Borner, T., and Wilde, A. (2005) Phototaxis in the cyanobac-teriumSynechocystis sp. PCC6803. Role of different photoreceptors.Pho-tochem. Photobiol. 81, 1481–1488

4. Ntefidou,M., Iseki,M.,Watanabe,M., Lebert,M., andHader, D.-P. (2003)Photoactivated adenylyl cyclase controls phototaxis in the flagellate Eu-glena gracilis. Plant Physiol. 133, 1517–1521

5. Masuda, S., and Bauer, C. E. (2002)AppA is a blue light photoreceptor thatantirepresses photosynthesis gene expression inRhodobacter sphaeroides.Cell 110, 613–623

6. Braatsch, S., Gomelsky, M., Kuphal, S., and Klug, G. (2002) A single flavo-protein, AppA, integrates both redox and light signals in Rhodobactersphaeroides.Mol. Microbiol. 45, 827–836

7. Tschowri, N., Busse, S., and Hengge, R. (2009) The BLUF-EAL proteinYcgF acts as a direct anti-repressor in a blue light response of Escherichiacoli. Genes Dev. 23, 522–534

8. Mussi, M. A., Gaddy, J. A., Cabruja, M., Arivett, B. A., Viale, A. M., Rasia,R., and Actis, L. A. (2010) The opportunistic human pathogen Acineto-bacter baumannii senses and responds to light. J. Bacteriol. 192,6336–6345

9. Stierl, M., Stumpf, P., Udwari, D., Gueta, R., Hagedorn, R., Losi, A., Gart-ner, W., Petereit, L., Efetova, M., Schwarzel, M., Oertner, T. G., Nagel, G.,and Hegemann, P. (2011) Light modulation of cellular cAMP by a smallbacterial photoactivated adenylyl cyclase, bPAC, of the soil bacteriumBeggiatoa. J. Biol. Chem. 286, 1181–1188

10. Weissenberger, S., Schultheis, C., Liewald, J. F., Erbguth, K., Nagel, G., andGottschalk, A. (2011) PAC�. An optogenetic tool for in vivomanipulationof cellular cAMP levels, neurotransmitter release, and behavior inCaenorhabditis elegans. J. Neurochem. 116, 616–625

11. Nagahama, T., Suzuki, T., Yoshikawa, S., and Iseki, M. (2007) Functionaltransplant of photoactivated adenylyl cyclase (PAC) into Aplysia sensoryneurons. Neurosci. Res. 59, 81–88

12. Bucher, D., and Buchner, E. (2009) Stimulating PAC� increases miniatureexcitatory junction potential frequency at the Drosophila neuromuscularjunction. J. Neurogenet 23, 220–224

13. Gauden,M., van Stokkum, I. H., Key, J. M., Luhrs, D. C., vanGrondelle, R.,Hegemann, P., and Kennis, J. T. (2006) Hydrogen bond switching througha radical pair mechanism in a flavin-binding photoreceptor. Proc. Natl.Acad. Sci. U.S.A. 103, 10895–10900

14. Bonetti, C., Mathes, T., van Stokkum, I. H., Mullen, K. M., Groot, M. L.,van Grondelle, R., Hegemann, P., and Kennis, J. T. (2008) Hydrogen bondswitching among flavin and amino acid side chains in the BLUF photore-ceptor observed by ultrafast infrared spectroscopy. Biophys. J. 95,4790–4802

15. Mathes, T., Zhu, J., van Stokkum, I. H., Groot, M. L., Hegemann, P., andKennis, J. T. (2012) Hydrogen bond switching among flavin and aminoacids determines the nature of proton-coupled electron transfer in BLUFphotoreceptors. J. Phys. Chem. Lett. 3, 203–208

16. Kennis, J. T., and Groot, M. L. (2007) Ultrafast spectroscopy of biologicalphotoreceptors. Curr. Opin. Struct. Biol. 17, 623–630

17. Masuda, S., Hasegawa, K., Ishii, A., and Ono, T. A. (2004) Light-inducedstructural changes in a putative blue light receptor with a novel FADbinding fold sensor of blue light using FAD (BLUF). Slr1694 of Synechocys-tis sp. PCC6803. Biochemistry 43, 5304–5313

18. Hasegawa, K., Masuda, S., and Ono, T. A. (2004) Structural intermediatein the photocycle of a BLUF (sensor of blue light using FAD) proteinSlr1694 in aCyanobacterium Synechocystis sp. PCC6803.Biochemistry 43,14979–14986

19. Grinstead, J. S., Avila-Perez, M., Hellingwerf, K. J., Boelens, R., andKaptein, R. (2006) Light-induced flipping of a conserved glutamine sidechain and its orientation in the AppA BLUF domain. J. Am. Chem. Soc.128, 15066–15067

20. Unno, M., Sano, R., Masuda, S., Ono, T. A., and Yamauchi, S. (2005)Light-induced structural changes in the active site of the BLUF domain inAppA by Raman spectroscopy. J. Phys. Chem. B 109, 12620–12626

21. Iwata, T., Watanabe, A., Iseki, M., Watanabe, M., and Kandori, H. (2011)Strong donation of the hydrogen bond of tyrosine during photoactivationof the BLUF domain. J. Phys. Chem. Lett. 2, 1015–1019

22. Barends, T. R., Hartmann, E., Griese, J. J., Beitlich, T., Kirienko, N. V.,Ryjenkov, D. A., Reinstein, J., Shoeman, R. L., Gomelsky,M., and Schlicht-ing, I. (2009) Structure andmechanism of a bacterial light-regulated cyclicnucleotide phosphodiesterase. Nature 459, 1015–1018

23. Yuan, H., Anderson, S., Masuda, S., Dragnea, V., Moffat, K., and Bauer, C.(2006) Crystal structures of the Synechocystis photoreceptor Slr1694 re-veal distinct structural states related to signaling. Biochemistry 45,12687–12694

24. Jung, A., Reinstein, J., Domratcheva, T., Shoeman, R., and Schlichting, I.(2006) Crystal structures of the AppA BLUF domain photoreceptor pro-vide insights into blue light-mediated signal transduction. J. Mol. Biol.362, 717–732

25. Kita, A.,Okajima, K.,Morimoto, Y., Ikeuchi,M., andMiki, K. (2005) Struc-ture of a cyanobacterial BLUF protein, Tll0078, containing a novel FAD-binding blue light sensor domain. J. Mol. Biol. 349, 1–9

26. Jung, A., Domratcheva, T., Tarutina, M., Wu, Q., Ko, W. H., Shoeman,R. L., Gomelsky,M., Gardner, K. H., and Schlichting, I. (2005) Structure ofa bacterial BLUF photoreceptor. Insights into blue light-mediated signaltransduction. Proc. Natl. Acad. Sci. U. S. A. 102, 12350–12355

27. Anderson, S., Dragnea, V., Masuda, S., Ybe, J., Moffat, K., and Bauer, C.(2005) Structure of a novel photoreceptor, the BLUF domain of AppAfrom Rhodobacter sphaeroides. Biochemistry 44, 7998–8005

28. Sadeghian, K., Bocola,M., and Schutz,M. (2008) A conclusivemechanismof the photoinduced reaction cascade in blue light using flavin photore-ceptors. J. Am. Chem. Soc. 130, 12501–12513

29. Domratcheva, T., Grigorenko, B. L., Schlichting, I., and Nemukhin, A. V.(2008) Molecular models predict light-induced glutamine tautomeriza-tion in BLUF photoreceptors. Biophys. J. 94, 3872–3879

30. Stelling, A. L., Ronayne, K. L., Nappa, J., Tonge, P. J., and Meech, S. R.(2007) Ultrafast structural dynamics in BLUF domains. Transient infraredspectroscopy of AppA and its mutants. J. Am. Chem. Soc. 129,15556–15564

31. Udvarhelyi, A., and Domratcheva, T. (2011) Photoreaction in BLUF re-ceptors. Proton-coupled electron transfer in the flavin-Gln-Tyr system.Photochem. Photobiol. 87, 554–563

32. Moglich, A., Yang, X., Ayers, R. A., and Moffat, K. (2010) Structure andfunction of plant photoreceptors. Annu. Rev. Plant Biol. 61, 21–47

33. Barry, B. A. (2011) Proton-coupled electron transfer and redox active ty-rosines in photosystem II. J. Photochem. Photobiol. B 104, 60–71

34. Huynh, M. H., and Meyer, T. J. (2007) Proton-coupled electron transfer.Chem. Rev. 107, 5004–5064

35. Moore, G. F., Hambourger, M., Kodis, G., Michl, W., Gust, D., Moore,T. A., and Moore, A. L. (2010) Effects of protonation state on a tyrosine-histidine bioinspired redox mediator. J. Phys. Chem. B 114, 14450–14457

36. Mathes, T., van Stokkum, I. H., Bonetti, C., Hegemann, P., andKennis, J. T.(2011) The hydrogen bond switch reaction of the Blrb Bluf domain ofRhodobacter sphaeroides. J. Phys. Chem. B 115, 7963–7971

37. Gauden, M., Yeremenko, S., Laan, W., van Stokkum, I. H., Ihalainen, J. A.,van Grondelle, R., Hellingwerf, K. J., and Kennis, J. T. (2005) Photocycle ofthe flavin-binding photoreceptor AppA, a bacterial transcriptional antire-pressor of photosynthesis genes. Biochemistry 44, 3653–3662

38. Zirak, P., Penzkofer, A., Schiereis, T., Hegemann, P., Jung, A., andSchlichting, I. (2005) Absorption and fluorescence spectroscopic charac-terization of BLUFdomain ofAppA fromRhodobacter sphaeroides.Chem.Phys. 315, 142–154

39. Zirak, P., Penzkofer, A., Hegemann, P., and Mathes, T. (2007) Photody-namics of BLUF domain mutant H44R of AppA from Rhodobacter spha-eroides. Chemical Physics. Chem. Phys. 335, 15–27

40. Nunthaboot, N., Tanaka, F., and Kokpol, S. (2009) Analysis of photoin-duced electron transfer inAppA. J. Photochem. Photobiol. A 207, 274–281

41. Dragnea, V., Waegele, M., Balascuta, S., Bauer, C., and Dragnea, B. (2005)Time-resolved spectroscopic studies of the AppA blue light receptor




ww

.jbc.org/D

ownloaded from

http://www.jbc.org/

BLUF domain from Rhodobacter sphaeroides. Biochemistry 44,15978–15985

42. Shibata, Y., Murai, Y., Satoh, Y., Fukushima, Y., Okajima, K., Ikeuchi, M.,and Itoh, S. (2009) Acceleration of electron transfer-induced fluorescencequenching upon conversion to the signaling state in the blue-light recep-tor, TePixD, from Thermosynechococcus elongatus. J. Phys. Chem. B 113,8192–8198

43. Gauden,M., Grinstead, J. S., Laan,W., van Stokkum, I. H., Avila-Perez,M.,Toh, K. C., Boelens, R., Kaptein, R., van Grondelle, R., Hellingwerf, K. J.,and Kennis, J. T. (2007) On the role of aromatic side chains in the photo-activation of BLUF domains. Biochemistry 46, 7405–7415

44. Bonetti, C., Stierl, M., Mathes, T., van Stokkum, I. H., Mullen, K. M.,Cohen-Stuart, T. A., van Grondelle, R., Hegemann, P., and Kennis, J. T.(2009) The role of key amino acids in the photoactivation pathway of theSynechocystis Slr1694 BLUF domain. Biochemistry 48, 11458–11469

45. Ishikita, H. (2007) Influence of the protein environment on the redoxpotentials of flavodoxins fromClostridium beijerinckii. J. Biol. Chem. 282,25240–25246

46. Ishikita, H. (2008) Light-induced hydrogen bonding pattern and drivingforce of electron transfer in AppA BLUF domain photoreceptor. J. Biol.Chem. 283, 30618–30623

47. Mathes, T., Vogl, C., Stolz, J., and Hegemann, P. (2009) In vivo generationof flavoproteins with modified cofactors. J. Mol. Biol. 385, 1511–1518

48. Datta, S., Costantino, N., and Court, D. L. (2006) A set of recombineeringplasmids for gram-negative bacteria. Gene 379, 109–115

49. Datsenko, K. A., and Wanner, B. L. (2000) One-step inactivation of chro-mosomal genes in Escherichia coli K-12 using PCR products. Proc. Natl.Acad. Sci. U.S.A. 97, 6640–6645

50. Mathes, T. (2007) Photochemie und Signaltransduktion von Blaulichtrezep-torproteinen aus photosynthetisierendenMikroorganismen. pp. 62–66,Hum-boldt University Berlin, Berlin

51. Bordoli, L., Kiefer, F., Arnold, K., Benkert, P., Battey, J., and Schwede, T.(2008) Nat. Protocols 4, 1–13

52. Berera, R., van Grondelle, R., and Kennis, J. T. (2009) Ultrafast transientabsorption spectroscopy. Principles and application to photosyntheticsystems. Photosynth. Res. 101, 105–118

53. Larsen, D. S., van Stokkum, I. H., Vengris, M., van Der Horst, M. A., deWeerd, F. L., Hellingwerf, K. J., and van Grondelle, R. (2004) Incoherentmanipulation of the photoactive yellow protein photocycle with dispersedpump-dump-probe spectroscopy. Biophys. J. 87, 1858–1872

54. Mullen, K. M., and van Stokkum, I. H. (2007) TIMP: An R Package forModeling Multi-way Spectroscopic Measurements. J. Stat. Softw. 18,1–46

55. Snellenburg, J. J., Laptenok, S. P., Seger, R., Mullen, K. M., and van Stok-kum, I. H. (2012) Glotaran. A Java-based graphical user interface for the Rpackage TIMP. J. Stat. Softw. 49, 1–22

56. Laan, W., Bednarz, T., Heberle, J., and Hellingwerf, K. J. (2004) Chro-mophore composition of a heterologously expressed BLUF domain. Pho-tochem. Photobiol. Sci. 3, 1011

57. Miura, R. (2001) Versatility and specificity in flavoenzymes. Controlmechanisms of flavin reactivity. Chem. Rec. 1, 183–194

58. van den Berg, P. A., van Hoek, A., and Visser, A. J. (2004) Evidence for anovel mechanism of time-resolved flavin fluorescence depolarization inglutathione reductase. Biophys. J. 87, 2577–2586

59. Seyedsayamdost, M. R., Reece, S. Y., Nocera, D. G., and Stubbe, J. (2006)Mono-, di-, tri-, and tetra-substituted fluorotyrosines. New probes forenzymes that use tyrosyl radicals in catalysis. J. Am. Chem. Soc. 128,1569–1579

60. Ghisla, S., and Massey, V. (1989) Mechanisms of flavoprotein-catalyzedreactions. Eur. J. Biochem. 181, 1–17

61. Fraaije, M. W., and Mattevi, A. (2000) Flavoenzymes. Diverse catalystswith recurrent features. Trends Biochem. Sci. 25, 126–132

62. Arents, J. C., Perez,M.A., Hendriks, J., andHellingwerf, K. J. (2011)On themidpoint potential of the FAD chromophore in a BLUF domain-contain-ing photoreceptor protein. FEBS Lett. 585, 167–172

63. Grinstead, J. S., Hsu, S. T., Laan, W., Bonvin, A. M., Hellingwerf, K. J.,Boelens, R., and Kaptein, R. (2006) The solution structure of the AppA

BLUF domain. Insight into the mechanism of light-induced signaling.Chembiochem 7, 187–193

64. Marcus, R. A., and Sutin, N. (1985) Electron transfer in chemistry andbiology. Biochim. Biophys. Acta 811, 265–322

65. Grampp, G. (1993) The Marcus inverted region from theory to experi-ment. Angew. Chem. Int. Ed. Engl. 32, 691–693

66. Thanasekaran, P., Rajendran, T., Rajagopal, S., Srinivasan, C., Ramaraj, R.,Ramamurthy, P., and Venkatachalapathy, B. (1997) Marcus inverted re-gion in the photoinduced electron transfer reactions of ruthenium (II)-polypyridine complexes with phenolate ions. J. Phys. Chem. A 101,8195–8199

67. Porcal, G., Bertolotti, S. G., Previtali, C. M., and Encinas, M. V. (2003)Electron transfer quenching of singlet and triplet excited states of flavinsand lumichrome by aromatic and aliphatic electron donors. Phys. Chem.Chem. Phys. 5, 4123–4128

68. Moser, C. C., Page, C. C., Farid, R., and Dutton, P. L. (1995) Biologicalelectron transfer. J. Bioenerg. Biomembr. 27, 263–274

69. Liu, Z., Guo, X., Tan, C., Li, J., Kao, Y. T.,Wang, L., Sancar, A., and Zhong,D. (2012) Electron tunneling pathways and role of adenine in repair ofcyclobutane pyrimidine dimer byDNAphotolyase. J. Am. Chem. Soc. 134,8104–8114

70. Pal, P. P., Bae, J. H., Azim, M. K., Hess, P., Friedrich, R., Huber, R., Mo-roder, L., and Budisa, N. (2005) Structural and spectral response of Ae-quorea victoria green fluorescent proteins to chromophore fluorination.Biochemistry 44, 3663–3672

71. Miroux, B., andWalker, J. E. (1996) Overproduction of proteins in Esche-richia coli. Mutant hosts that allow synthesis of some membrane proteinsand globular proteins at high levels. J. Mol. Biol. 260, 289–298

72. Dunitz, J. D. (2004) Organic fluorine. Odd man out. ChemBioChem 5,614–621

73. Toh, K. C., van Stokkum, I. H., Hendriks, J., Alexandre,M. T., Arents, J. C.,Perez, M. A., van Grondelle, R., Hellingwerf, K. J., and Kennis, J. T. (2008)On the signaling mechanism and the absence of photoreversibility in theAppA BLUF domain. Biophys. J. 95, 312–321

74. Weber, S., Schroeder, C., Kacprzak, S., Mathes, T., Kowalczyk, R. M.,Essen, L. O., Hegemann, P., Schleicher, E., and Bittl, R. (2011) Light-gen-erated paramagnetic intermediates in BLUF domains. Photochem. Photo-biol. 87, 574–583

75. Zirak, P., Penzkofer, A., Schiereis, T., Hegemann, P., Jung, A., andSchlichting, I. (2006) Photodynamics of the small BLUF protein BlrB fromRhodobacter sphaeroides. J. Photochem. Photobiol. B 83, 180–194

76. Masuda, S., Tomida, Y., Ohta, H., and Takamiya, K. I. (2007) The criticalrole of a hydrogen bond between Gln-63 and Trp-104 in the blue light-sensing BLUF domain that controls AppA activity. J. Mol. Biol. 368,1223–1230

77. van der Horst, M. A., Laan, W., Yeremenko, S., Wende, A., Palm, P.,Oesterhelt, D., and Hellingwerf, K. J. (2005) From primary photochemis-try to biological function in the blue light photoreceptors PYP and AppA.Photochem. Photobiol. Sci. 4, 688–693

78. Jager, A., Braatsch, S., Haberzettl, K., Metz, S., Osterloh, L., Han, Y.,and Klug, G. (2007) The AppA and PpsR proteins from Rhodobactersphaeroides can establish a redox-dependent signal chain but fail totransmit blue light signals in other bacteria. J. Bacteriol. 189,2274–2282

79. Braatsch, S.,Moskvin, O. V., Klug, G., andGomelsky,M. (2004) Responsesof the Rhodobacter sphaeroides transcriptome to blue light under semi-aerobic conditions. J. Bacteriol. 186, 7726–7735

80. Metz, S., Jager, A., and Klug, G. (2009) In vivo sensitivity of blue light-de-pendent signaling mediated by AppA/PpsR or PrrB/PrrA in Rhodobactersphaeroides. J. Bacteriol. 191, 4473–4477

81. Pandey, R., Flockerzi, D., Hauser, M. J., and Straube, R. (2011) Modelingthe light- and redox-dependent interaction of PpsR/AppA in Rhodobactersphaeroides. Biophys. J. 100, 2347–2355

82. Han, Y., Meyer, M. H., Keusgen, M., and Klug, G. (2007) A heme cofactoris required for redox and light signaling by the AppA protein of Rhodo-bacter sphaeroides.Mol. Microbiol. 64, 1090–1104




ww

.jbc.org/D

ownloaded from

http://www.jbc.org/

Tilo Mathes, Ivo H. M. van Stokkum, Manuela Stierl and John T. M. KennisProton-coupled Electron Transfer and Photoactivation of BLUF Photoreceptors

Redox Modulation of Flavin and Tyrosine Determines Photoinduced

doi: 10.1074/jbc.M112.391896 originally published online July 25, 20122012, 287:31725-31738.J. Biol. Chem.

10.1074/jbc.M112.391896Access the most updated version of this article at doi:

Alerts:

When a correction for this article is posted•

When this article is cited•

to choose from all of JBC's e-mail alertsClick here

Supplemental material:

http://www.jbc.org/content/suppl/2012/07/25/M112.391896.DC1

http://www.jbc.org/content/287/38/31725.full.html#ref-list-1

This article cites 81 references, 12 of which can be accessed free at


ww

.jbc.org/D

ownloaded from

http://www.jbc.org/lookup/doi/10.1074/jbc.M112.391896

http://www.jbc.org/cgi/alerts?alertType=citedby&addAlert=cited_by&cited_by_criteria_resid=jbc;287/38/31725&saveAlert=no&return-type=article&return_url=http://www.jbc.org/content/287/38/31725

http://www.jbc.org/cgi/alerts?alertType=correction&addAlert=correction&correction_criteria_value=287/38/31725&saveAlert=no&return-type=article&return_url=http://www.jbc.org/content/287/38/31725

http://www.jbc.org/cgi/alerts/etoc

http://www.jbc.org/content/suppl/2012/07/25/M112.391896.DC1

http://www.jbc.org/content/287/38/31725.full.html#ref-list-1

http://www.jbc.org/